2012美国大学生数学建模答案.docVIP

Modeling Overnight Recreation Trip Choice: Application of a Repeated Nested Multinomial Logit Model ? W. DOUGLASS SHAW1 and MICHAEL T. OZOG2 1Department of Applied Economics and Statistics/204, University of Nevada, Reno, NV 89557-0105, USA (e-mail: wdshaw@unr.edu); 2Quantitative Research Group, 1712 Westchester Lane, Fort Collins, CO 80525, USA Accepted 8 April 1998 Abstract. In this paper we apply the repeated nested multinomial logit model, a version of a random utility model (RUM), to estimate the choice of an overnight versus single day recreation trip, along with the other usual choice of which of the sites to visit, and less typically, the choice of whether to participate (in our application – to fish) at all. We also find statistically significant income effects in the empirical results. The application is to Atlantic Salmon fishing and the data set is for Maine resident angler’s fishing trips to rivers in Maine and Canada. Key words: repeated nested multinomial logit, RUM, recreation demand, salmon fishing, trip length decisions JEL Classification: C25, D61, Q26 1. Introduction In this paper we apply the repeated nested multinomial logit model, a version of a random utility model (RUM), to estimate the choice of an overnight versus single day recreation trip, along with the other usual choice of which of the sites to visit, and less typically, the choice of whether to participate (in our application – to fish) at all. A sample of recreators can include individuals who are more likely to take a multiple day visit rather than a single or part day visit. However, even when they appear in the original sample of those who return the survey, such multiple day trip takers are often omitted from the estimating sample for purposes of predicting demand and calculating consumer’s surplus. Omitting them from the sample understates the total value (i.e. aggregated across the population who may visit the sites) for recreational resources, presuming these individuals have